sensitivity of cancer cells to truncated diphtheria toxin...cam tumor growth assay hep3-gfp tumour...
TRANSCRIPT
Sensitivity of Cancer Cells to Truncated Diphtheria ToxinYi Zhang1, Wendy Schulte2, Desmond Pink2, Kyle Phipps1, Andries Zijlstra2,3, John D. Lewis2,4, David
Morton Waisman1*
1 Departments of Biochemistry and Molecular Biology and Pathology, Dalhousie University, Halifax, Nova Scotia, Canada, 2 Innovascreen Inc, Halifax, Nova Scotia, Canada,
3 Department of Pathology, Vanderbilt University, Nashville, Tennessee, Unites States of America, 4 Department of Oncology, University of Western Ontario, London,
Ontario, Canada
Abstract
Background: Diphtheria toxin (DT) has been utilized as a prospective anti-cancer agent for the targeted delivery of cytotoxictherapy to otherwise untreatable neoplasia. DT is an extremely potent toxin for which the entry of a single molecule into acell can be lethal. DT has been targeted to cancer cells by deleting the cell receptor-binding domain and combining theremaining catalytic portion with targeting proteins that selectively bind to the surface of cancer cells. It has been assumedthat ‘‘receptorless’’ DT cannot bind to and kill cells. In the present study, we report that ‘‘receptorless’’ recombinant DT385 isin fact cytotoxic to a variety of cancer cell lines.
Methods: In vitro cytotoxicity of DT385 was measured by cell proliferation, cell staining and apoptosis assays. For in vivostudies, the chick chorioallantoic membrane (CAM) system was used to evaluate the effect of DT385 on angiogenesis. TheCAM and mouse model system was used to evaluate the effect of DT385 on HEp3 and Lewis lung carcinoma (LLC) tumorgrowth, respectively.
Results: Of 18 human cancer cell lines tested, 15 were affected by DT385 with IC50 ranging from 0.12–2.8 mM. Furthermore,high concentrations of DT385 failed to affect growth arrested cells. The cellular toxicity of DT385 was due to the inhibitionof protein synthesis and induction of apoptosis. In vivo, DT385 diminished angiogenesis and decreased tumor growth in theCAM system, and inhibited the subcutaneous growth of LLC tumors in mice.
Conclusion: DT385 possesses anti-angiogenic and anti-tumor activity and may have potential as a therapeutic agent.
Citation: Zhang Y, Schulte W, Pink D, Phipps K, Zijlstra A, et al. (2010) Sensitivity of Cancer Cells to Truncated Diphtheria Toxin. PLoS ONE 5(5): e10498.doi:10.1371/journal.pone.0010498
Editor: Joseph Alan Bauer, Bauer Research Foundation, United States of America
Received November 5, 2009; Accepted April 14, 2010; Published May 5, 2010
Copyright: � 2010 Zhang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by a grant from the Nova Scotia Health Research Foundation (NSHRF). YZ is supported by a studentship from the CancerResearch Training Program, Dalhousie University. JDL is supported by grant #018176 from the NCIC/Terry Fox Foundation. The funders had no role in studydesign, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: [email protected]
Introduction
Diphtheria toxin (DT) is synthesized in Corynebacterium diphtheriae
as a single-chain enzyme of 535 amino acids with a molecular
weight of 63,000 [1,2]. DT consists of three key domains: the
amino-terminal C, or catalytic, domain (residues 1–186); the
intermediate T, or transmembrane, domain (residues 202–381);
and the carboxyl-terminal R, or receptor-binding, domain
(residues 391–535). The catalytic domain is connected to the T
domain by an arginine-rich loop and a readily reducible disulfide
bridge (linking C186 to C201). DT has been shown to enter toxin-
sensitive mammalian cells by receptor-mediated endocytosis which
involves the interaction of the receptor-binding domain of the
protein with a transmembrane cell surface precursor of the
heparin-binding epidermal growth factor-like growth factor [3,4].
After binding to this cell-surface receptor, DT is endocytosed and
trafficked to an acidic vesicular compartment, where it undergoes
a pH-dependent conformational change, cleavage and release of
the catalytic domain. The T domain inserts into the vesicular
membrane and the resultant channel is utilized for the
translocation of the catalytic domain to the cytosol. There, the
catalytic subunit catalyzes the ADP-ribosylation of elongation
factor 2, resulting in the inhibition of protein synthesis and cell
death (reviewed in [5]).
A number of truncated, recombinant DT proteins have been
produced in which the receptor-binding domain has been
genetically replaced by ligands that can selectively target
malignant cells. These fusion proteins represent a novel class of
cytotoxic agents which, unlike chemotherapeuticDT has been
shown to enter toxin-sensitive mammalian cells by receptor-
mediated endocytosis which involves the interaction of the
receptor-binding domain of the protein with drugs, kill targeted
cells by inhibiting protein synthesis and thereby inducing
apoptosis[6]. These fusion proteins include DT508-MSF [7],
DT486-IL-2 [8], DT486-GM-CSF [9], DT390-IL3 [10], DT388-
GM-CSF [11–13], DT388-IL-3 [14,15], DT385-VEGF [16,17]
and DT388 combined with the ATF domain of uPA [18]. Among
the resulting drugs, DT388IL-3 has shown some promise in
clinical trials [19,20], whereas the DT389-IL-2 recombinant toxin
(DAB389-IL-2, denileukin diftitox-Ontak) has been approved by
the FDA for clinical use in advanced stage cutaneous T-cell
lymphoma (reviewed in [21–23].
PLoS ONE | www.plosone.org 1 May 2010 | Volume 5 | Issue 5 | e10498
It is widely accepted that the efficacy of the DT fusion proteins
lies in the ability of the targeting ligand component to direct the
DT to cancer cells resulting in targeted cellular toxicity.
Furthermore, the removal of the DT receptor-binding domain is
expected to result in a truncated DT that is unable to interact with
its receptor on the surface of eukaryotic cells and therefore unable
to bind to and kill cells. This concept has been reinforced by the
report that the truncated DT (DT385) is not cytotoxic [16].
In the current study, we show that contrary to previous reports,
the recombinant truncated DT, DT385 is cytotoxic to many
cancer cells. We also observed that DT385 inhibits the growth of
human and mouse tumors. Our findings establish the efficacy of
DT385 as a potential antitumor agent.
Materials and Methods
Cell LinesHuman Umbilical Vein Endothelial Cells (HUVEC) were
obtained from Cell Applications, Inc. and grown in an endothelial
cell growth medium with full growth supplements (Cell Applica-
tions, Inc.). Bovine pulmonary artery endothelial cells (BPAEC)
and human dermal microvascular endothelial cells (HDMEC)
were obtained from Lonza and were grown in an EBM medium
plus EGM SingleQuots of growth supplements and EBM-2
medium plus EGM-2 SingleQuots of growth supplements (Lonza),
respectively. Glioma cell lines U-87 MG and U251 were kindly
provide by Dr. V. Wee Yong (University of Calgary, Calgary,
Alberta, Canada). The human epidermoid carcinoma cell line
HEp3 was a generous gift from Dr. Andries Zijlstra (Vanderbilt
University, USA). Mouse embryonic fibroblast (MEF) cells were
isolated from mouse embryos and were used at their early passages
(less than passage 4). U-87 MG, U251, HEp3 and MEF cells were
cultured in DMEM containing 10% (v/v) fetal bovine serum (FBS,
Invitrogen) and 1% penicillin-streptomycin mixtures (Invitrogen).
All other cell lines were obtained from American Type Culture
Collection (ATCC, Rockville, MD) and were grown in DMEM,
MEM or RPMI (Invitrogen) containing 10% (v/v) fetal bovine
serum and 1% penicillin-streptomycin mixtures according to
ATCC’s instructions. All cells were grown in an incubator at 37uCcontaining 5% CO2. All endothelial cells and primary fibroblast
cells were maintained and used before the ninth passage.
Induction and purification of recombinant proteinsThe plasmid pET17b-DT385 expressing ‘‘receptorless’’ DT385
was generously provided by Dr. Sundaram Ramakrishnan
(Department of Pharmacology, University of Minnesota Medical
School, Minneapolis, MN). The plasmid pET17b-p22 was
constructed by cloning human plasminogen fragment p22 into
pET17b (EMD Biosciences) at the NdeI and BamHI restriction
sites. The plasmid pLIC-DT385-p22 was constructed by cloning
human plasminogen fragment p22 into the plasmid pLIC-DT385
at the NcoI and HindIII restriction sites, while the plasmid pLIC-
DT385 was constructed by inserting DNA encoding DT385 but
lacking the stop codon into the pET-30 EK/LIC vector following
the manufacturer’s protocol (Novagen). The plasmid pSUMO
encoding the cDNA for SUMO (small ubiquitin-related modifier,
Saccharomyces cerevisiae, Smt3 gene) was kindly provided by Dr.
Kaisong Zhou (Dalhousie University). All plasmid constructs were
confirmed by DNA sequencing (DalGEN, the Dalhousie Univer-
sity DNA sequencing facility). Recombinant p22, SUMO, DT385,
and DT-p22 were expressed in E. coli and purified by Ni-NTA
affinity column (Qiagen), pooled and dialyzed overnight against
phosphate buffered saline (PBS). All recombinant proteins were
purified as a single band as revealed by SDS-PAGE analysis. LPS
was removed from purified proteins using polymixin B beads
(Detoxi-Gel Endotoxin Removing Gel, Pierce) according to the
manufacturer’s instructions. Multiple passes through the column
were performed until LPS level was less than 30 EU/mg.
Cell viability assayCell viability was assessed by the Cell Titer96 Aqueous One
Solution Cell Proliferation Assay (MTS assay-Promega) using the
manufacturer’s protocol. The IC50 values were calculated by
nonlinear least-squares fitting of dose-response curves using the
open source computer program QTI PLOT. Data were analyzed
with the four-parameter logistic equation f = (a 2 d)/[1 + (x/c)b]
+ d, where a is the asymptotic maximum, b is a slope parameter, c
is the value at the inflection point (IC50) and d is the asymptotic
minimum. For a control, 0.4 mM or 1.2 mM DT385 was added to
the tissue culture media in the absence of cells followed by
incubation with the MTS/PMS reagent. Reduction of MTS was
not observed under these conditions, confirming that DT385 did
not interfere with this assay.
Crystal violet stainCell proliferation was also evaluated with crystal violet staining.
At the end of treatment with DT385, cells were fixed with 100%
methanol, and stained with 0.5% crystal violet in 20% methanol
for 15 minutes at room temperature. Stained cells were
photographed at 256 magnification on a Zeiss Axiover 200
inverted microscope (Carl Zeiss, Germany).
Apoptosis and Necrosis AssayApoptotic and necrotic cells were visualized with the apoptosis
and necrosis assay kit (Biotium, Inc) following the manufacturer
protocol. Stained cells were photographed on a Zeiss Axioplan II
fluorescence microscope (Carl Zeiss, Germany), with appropriate
filter settings for fluorescein isothiocyanate (FITC) and Texas Red
fluorescence. Digital images were processed in Adobe Photoshop
(Adobe Inc.).
Protein synthesis assayU-87 MG or Hela cells were seeded at a ratio of 1:10 in 1 ml of
DMEM plus 10% FBS per well in a 12-well plate overnight. Cells
were treated with 1 mM DT385 for the indicated time, incubated
for 15 minutes at 37uC with 0.5 ml labeling medium (methionine-
free DMEM, 10% dialyzed FBS and 50–100 mCi Pro-Mix L-[35-
S]-(Amersham)) and cell lysates analyzed by SDS-PAGE. Radio-
active bands were visualized by radiography using a phosphor-
Imager after overnight exposure on a phosphor screen. For
quantitation, [35S] incorporation was measured by liquid scintil-
lation counting.
Chick chorioallantoic membrane (CAM) angiogenesisassay
The anti-angiogenic activity of proteins was tested on the CAM
as described [24]. When analyzing tumor-induced vascularization,
50,000 HEp3 cells replaced bFGF/ VEGF as the angiogenic
stimulus [25]. Four implants were placed on each CAM and ten
embryos were used for each experimental compound.
CAM tumor growth assayHEp3-GFP tumour cells (100,000 cells/10 ml DMEM media)
were applied directly to a filter-disc abraded area of the CAM of 9-
day old chick embryos as detailed by [26]. Chicks were incubated
under standard conditions until Day 15 when tumors were imaged
and sorted for random distribution to control and treatment
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 2 May 2010 | Volume 5 | Issue 5 | e10498
groups. Daily systemic intravenous injections of DT385 (2 mg in
50 mL PBS) or PBS (50 mL) were performed on day 15, 16 and 17.
On day 18, tumors were excised, cleaned of CAM and then
weighed. Alternatively, fluorescent (GFP) Hep3 tumors were
imaged in vivo using a Zeiss Lumar fluorescence stereomicroscope.
The outline of the tumor was determined using the fluorescence
signal of the tumor cells (GFP channel (ex470/em525)). Specifi-
cally, Volocity software was calibrated to select areas with
fluorescence intensity corresponding to 1 or more standard
deviations above background. The area of the tumour was defined
and calculated using Improvision Volocity software (Perkin Elmer,
Inc.)(Version 5.3.0 Build 0). A unit of measure (1 mm) was
calibrated using the grid of a hemocytometer. The determination
of tumor volume assumes the shape to be hemiellipsoidal and the
equation for the tumor volume is: V = p /6 N (length) N (width) N(height) where height = 1.63!(Length NWidth). Data were analyzed
using Students t-test.
Mouse tumor modelAll animal work was carried out at the animal facility of
Dalhousie University in accordance with the guidelines set forth in
the Care and Use of Laboratory Animals by Dalhousie University.
All investigations were approved by the Dalhousie Animal
Research Ethics Board. Female 6-to 8-week old C57BL6/J mice
(Jackson Laboratories) were used. Tumors were induced by
subcutaneous injection of LLC cells (106 cells) in 100 ml of sterile
PBS. Palpable tumors were established three to four days after
injection, at which point the mice were randomly assigned to two
experimental groups, those that received recombinant SUMO
(control) and those that received recombinant DT385 treatment.
SUMO and DT385 were administered to mice, peritumorally at
day 5 (25 mg in 100 mL PBS), and at days 9, 12 and 15 (10 mg in
100 mL PBS each injection).
Protein Labeling with FITCDT385 and BSA were FITC-labeled using the EZ-label FITC
protein labeling Kit following the manufacturer’s protocol (Pierce).
Ammonium chloride washout studiesU87 cells were incubated in the absence or presence of DT385
(2 mM) alone or in combination with 10 mM ammonium chloride.
The medium was removed at various time points and replaced
with fresh media. Thirty six hours after the addition of test
compounds, cell viability was measured by the MTS assay.
Statistical AnalysisThe significance of the data was determined using the Student’s
t-test (one-tailed). P values of ,0.05 were regarded as significant.
Results
Characterization of DT385Recent studies from our laboratory identified and characterized
a novel antiangiogenic fragment of plasminogen called p22 [27].
Since p22 was a potent and specific inhibitor of capillary
endothelial cells, it was envisioned that this protein could be used
to target DT to newly forming vasculature. We expressed a fusion
protein in E. coli in which p22 was substituted for the receptor-
binding domain of DT (DT385-p22). The activity of this fusion
protein was compared with recombinant DT385 and recombinant
p22. To determine the impact of these compounds on cell viability,
MTS assays were performed on bovine pulmonary artery
endothelial cells (BPAEC) after a three-day treatment. Interest-
ingly, we observed that unlike the p22 prepared by proteolytic
digestion of human plasminogen [27], the recombinant p22 failed
to inhibit the viability of the BPAEC (Figure 1A). Unfortunately
we have been unable to produce active recombinant p22 in E. coli
or yeast (data not shown). Of particular interest was our
observation that both the recombinant DT385-p22 construct
and recombinant DT385 dramatically decreased the number of
viable BPAEC. We determined an IC50 value of 0.21 mM and
0.14 mM for DT385-p22 and DT385 respectively (Table 1).
Since the recombinant p22 was inactive whereas DT385-p22
retained activity, it was unclear from these results if p22 might
have retained activity when expressed as the fusion protein,
DT385-p22. Considering the specificity of plasminogen-derived
p22 for endothelial cells, we expected that if the p22 component of
the DT385-p22 fusion protein, then DT385-p22 should only
target endothelial cells and not cancer cells, as has been
demonstrated for plasminogen-derived p22 [27]. We therefore
tested the cytotoxic activity of DT385-p22 and DT385 with cancer
cell lines. For these experiments we initially tested two commonly
used human cancer cell lines, the HT1080 fibrosarcoma cell and
the HeLa cervical carcinoma cell. Unexpectedly, we found that
DT385 caused a dramatic loss in cell viability of the cancer cell
lines in a dose-dependent manner (Figure 1B). At the highest dose
tested (2.4 mM), 10% of viable Hela cells remained, while about
50% of viable HT1080 cells remained. DT385-p22 did also cause
a loss in cell viability but was less potent than DT385. To
determine whether the viability of human endothelial cells was
affected by DT385-p22, we incubated HUVEC and HDMEC
with these proteins (Figure 1C). Surprisingly, the viability of both
cell lines was not affected by DT385, D385-p22 or p22 at the
concentration up to 2.5 mM. We extended the incubation time to
7 days with 2.4 mM of these recombinant proteins, and no loss of
cell viability was observed (Figure S1,A). However, with very high
concentrations of DT385, loss of viability of the HUVEC and
HDMEC was observed (Figure S1,B). We determined an IC50 of
about 5.77 and 7.54 mM DT385 for the HUVEC and HDMEC,
respectively. These results suggested that the cytotoxic activity of
DT385-p22 was due to the activity of the DT385 domain of the
fusion protein and not the p22 domain. We also concluded that
DT385 can kill human cancer cells at a concentration not affecting
primary endothelial cell lines. The observation that DT385 had
cytotoxic activity was novel and unexpected. We therefore
investigated the possibility that DT385 might kill other cancer
cells.
Cytotoxic effects of DT385 on cancer cell linesWe extended the cytotoxicity assay to 18 human cancer cell
lines (Table 1). Fifteen cancer cell lines were affected by DT385
and efficacy of DT385 varied among the cancer cells. Cancer cell
lines with highest sensitivity to DT385 (IC50 less than 0.5 mM
DT385), included U-87 MG, U251, 293T, HEK293, Hela and
Calu-3 cells. The group with intermediate sensitivity to DT385
(IC50 between 0.5-1.5 mM) included Colo201, Colo205, LNCap,
PC-3, HT1080, and MDA-MB-231 cells. Weakly sensitive cancer
cell lines with IC50 greater than 1.5 mM included MCF7,
HCT116, BT-20, NB4, HL-60, and HEp3 cells. In contrast,
recombinant p22 had no effect on cell viability at concentrations
as high as 2.4 mM (data not shown). The three human cancer cell
lines not affected by DT385 at the highest dose tested (2.4 mM)
were the promyelocytic leukemia cell lines (HL-60 and NB4) and
the epidemoid carcinoma cell line (HEp3).
The DT385-mediated loss in cell viability was also confirmed
using crystal violet staining (Figure 2). For example, less than 10%
of glioma U-87 MG cells remained after DT385 treatment. To
confirm cross-species sensitivity to DT385, we tested mouse skin
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 3 May 2010 | Volume 5 | Issue 5 | e10498
melanoma (B16-F10), Lewis lung carcinoma (LLC) and mouse
embryonic fibroblast (MEF) cells. While the B16-F10 and MEF
were highly sensitive to DT385, the LLC were only weakly
sensitive (Table 1). Together the data from MTS assays and
Crystal Violet staining indicate that the ‘‘receptorless’’ DT,
DT385, can indeed be cytotoxic to many tumor cell lines. The
molecular mechanism that accounts for the broad range in
sensitivity to DT385 is currently under investigation.
Cytotoxic effects of DT385 on other primary cell linesTo evaluate the cytotoxicity of DT385 on other primary cell
lines, primary cultures of three human fibroblast cell lines (CCD-
1064Sk, BJ, and IMR-90) and one monocyte (SC) cell line were
incubated with increasing concentrations of DT385 and the cell
viability was determined by the MTS assay. CCD-1064Sk and BJ
fibroblasts had an IC50 of 2.22 mM and 2.37 mM DT385,
respectively. The lung fibroblasts (IMR-90) had an IC50 of
1.44 mM and the viability of the monocytes (SC) was unaffected by
DT385 at concentrations up to 2.4 mM (Table 1). Again,
recombinant p22 had no effect on cell viability at concentrations
up to 2.4 mM (data not shown). In contrast, when confluent CCD-
1064Sk, BJ and IMR-90 fibroblasts were treated for 3 days with
2 mM DT385, we observed no significant loss of cell viability
(Figure S2). These data indicate that even when DT385 is effective
Figure 1. Effect of DT385 fusion protein and DT385 on cellular viability. Recombinant p22, (expressed from pET17b-p22), DT385 (expressedfrom pET17b-DT385), DT385-p22 (expressed from pLIC-DT385-p22), (see Materials and Methods) or an equivalent volume of PBS (control) wereincubated with (A) BPAEC, (B) Hela and HT1080 cells, and (C) HUVEC and HDMEC cells in growth media for three days. After the three day incubation,viable cell numbers were quantified by the MTS assay. The PBS vehicle control was considered as 100% viable. Results are expressed as percentagesof PBS-treated cells. Results are the mean 6 S.D. of 3 independent experiments performed in triplicate (n = 9).doi:10.1371/journal.pone.0010498.g001
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 4 May 2010 | Volume 5 | Issue 5 | e10498
in decreasing the viability of proliferating primary cells, it fails to
kill these cells when they are quiescent/confluent. Taken together,
these data support our previous conclusion that DT385 alone has
the potential to kill cancer cells with minimum effects on primary
cells.
To determine if the resistance to DT385 could be overcome by
continuous incubation, cells with weak or no detectable sensitivity
received a second dose, 48 hours after the first treatment and cell
viability was measured 2 days later. The human epidermoid
carcinoma cell line, HEp3 and breast cancer cell line MDA-MB-
Table 1. Cytotoxicity of recombinant DT385.
Cell lines IC50 (mM) Potency Cell doubling time (h)
Human tumor or transformed cell lines
U-87 MG 0.3860.009 strong 27 ref. 34
U251 0.4660.0415 strong 40 M
293T 0.1260.027 strong 22 ref. 33
HEK293 0.1260.0085 strong 35 M
Hela 0.3360.013 strong 32 M
Calu-3 0.1360.022 strong 24 M
Colo201 0.8660.068 intermediate
Colo205 0.8760.059 intermediate
LNCap 0.9560.077 intermediate
PC-3 0.9860.11 intermediate 24M/18 ref. 32
HT1080 1.2160.09 intermediate 57 M
MDA-MB-231 (1 treatment) 1.0260.034 intermediate 43 M
MDA-MB-231 (2 treatments) 0.6660.024 intermediate 43 M
MCF7 2.2660.13 weak 16 M
HCT116 2.8260.28 weak 17.7 ref. 31
NB4 ND weak
BT-20 ND weak
HL-60 ND weak
TIME ND weak
HEp3 (1 treatment) ND weak
HEp3 (2 treatments) 2.0760.25 weak
Human primary cell lines
SC ND weak
CCD-1064Sk 2.2260.18 weak
BJ 2.3760.26 weak
IMR-90 1.4460.19 intermediate
HUVEC 5.7760.312 weak 27 ref. 35
HDMEC 7.5460.187 weak
Bovine cell line
BPAEC 0.13560.0076 strong
Mouse cell lines
MEF 0.3360.011 strong
B16F10 0.46360.011 strong 39 M
LLC 5.4260.29 weak
Cells were treated with varying concentrations of DT385 for 3 days and cell viability was quantified by the CellTiter 96 Aqueous (Promega) assay as described in thelegend to Figure 1. In order to determine the effect of double DT385 treatment, cells were treated with varying concentrations of DT385 for 2 days, washed with PBSand the procedure repeated for aCells were treated with varying concentrations of DT385 for 3 days and cell viability was quantified by the CellTiter 96 Aqueous(Promega) assay as described in the legend to Figure 1. In order to determine the effect of double DT385 treatment, cells were treated with varying concentrations ofDT385 for 2 days, washed with PBS and the procedure repeated for another 2 days (2 treatments). To allow comparison of single and double treatments, one group ofcells were incubated with varying concentrations of DT385 for 4 days (1 treatment).IC50 is the concentration of DT385 reducing cell viability to 50% after 3 days. Strong: IC50 is less than 0.5 mM; intermediate: IC50 is between 0.5 mM–1.5 mM; weak: IC50 isgreater than 1.5 mM.ND: Not determinable, the inhibition of cell viability by compounds was less than 20% at the concentration of 2.4 mM.M: measured in this study. Cells were grown in 96-well plates and measured daily by using the CellTiter 96 AQueous (Promega) assay. The doubling time of the cells wascalculated from the exponential portion of the growth curve. nother 2 days (2 treatments). To allow comparison of single and double treatments, one group of cellswere incubated with varying concentrations of DT385 for 4 days (1 treatment).doi:10.1371/journal.pone.0010498.t001
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 5 May 2010 | Volume 5 | Issue 5 | e10498
231 both exhibited enhanced sensitivity with continuous incuba-
tion with DT385. Compared to a single application of DT385,
which did not affect HEp3 cells significantly, two applications of
DT385 decreased the cell viability (IC50 of 2.07 mM (Figure S3).
Similarly, the IC50 for the breast cancer cell line, MDA-MB-231
decreased from 1.02 mM to 0.66 mM upon a second incubation
with DT385 (Table 1). In contrast, a second incubation with
DT385 was without effect for the primary endothelial cells,
HUVEC or HDMEC (Figure S3). Taken together, the results
show that receptorless DT in the form of DT385 is cytotoxic to a
large number of tumor cells. While the sensitivity to DT385 varies,
continuous exposure diminishes the viability of even the most
resistant cancer cell lines.
Mechanism of action of DT385DT kills cells by a mechanism involving cellular apoptosis. We
also observed increased apoptosis in DT385 treated cells (Figure 3).
The incubation of cultured U-87 with DT385 resulted in a
Figure 2. Cytotoxic effects of DT385 on cancer cell lines. Various cancer cell lines were cultured with 2 mM DT385 or recombinant p22 (rP22)for 3 days. Cells were then stained with crystal violet as described in Materials and Methods and images were obtained at 256magnification with aZeiss Axiover 200 inverted microscope (Carl Zeiss, Germany).doi:10.1371/journal.pone.0010498.g002
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 6 May 2010 | Volume 5 | Issue 5 | e10498
dramatic increase in cellular apoptosis as measured by an increase
in annexin V staining (85%) and the uptake of ethidium
homodimer (87%).
DT has been shown to enter toxin-sensitive mammalian cells by
receptor-mediated endocytosis which involves the interaction of
the receptor-binding domain of the protein with its extracellular
receptor. Since DT385 does not possess a receptor-binding
domain, it should be incapable of binding to cells and becoming
internalized. To investigate if DT385 was internalized, we tracked
fluorescently labeled DT385 with confocal microscopy. Tumor
cells sensitive to DT385 were incubated with FITC-DT385 and
imaged with confocal microscopy. As shown in Figure 4A,
incubation of cells with DT385 resulted in the internalization of
the toxin which was detectable in perinuclear vesicles. In contrast,
incubation of cells with similar concentrations of FITC-bovine
serum albumin did not result in internalization of the protein
(Figure S4). This experiment suggested that the uptake of DT385
by cells was not due to non specific cellular uptake of the protein.
Ammonium chloride is a weak base that diffuses into the
endosome and serves as a proton reservoir, thus inhibiting the
acidification of the endosome. We utilized ammonium chloride
treatment of cells to investigate whether DT385 entered the cell by
endocytosis. We observed that incubation of U87 cells with
DT385 for as little as 2 hours resulted in significant cell death
(Figure 4B). However, the simultaneous addition of ammonium
chloride and DT385 abrogated the cytotoxic activity of DT385.
The observation that ammonium chloride blocked the cytotoxic
activity of DT385 suggests that DT-385 enters cells through an
acidic endocytic pathway.
Diphtheria toxin kills cells by catalysing the ADP-ribosylation of
EF-2, leading to inhibition of protein synthesis [5,28]. To
investigate if the DT385 decreased cell viability by inhibiting
protein synthesis, glioma U-87 MG cells were treated with 1 mM
DT385 for 36 hours followed by labeling with [35S-] methionine
for 15 minutes. As shown in Figure 5A, SDS-PAGE analysis of cell
lysate showed that protein synthesis was severely reduced in
DT385 treated cells. This inhibition occurred within 24 hrs of
treatment and persisted during the duration of the assay (48 hr)
(Figure 5B). The 36-h treatment gave the greatest decrease in
protein labeling. These data suggest that, mechanistically, DT385
Figure 3. DT385 causes apoptosis and cell death in U-87 MG cells. (A), U-87 MG cells growing in an 8- well chamber slide were treated with1.2 mM DT385, or control (PBS) respectively for 3 days. Following treatments, cells were stained for apoptosis with FITC-labeled annexin V. Cellmembrane integrity was evaluated with ethidium homodimer III (EtD-III). Representative images (1006 magnifications) are shown. The arrowindicates cells that were labeled with both fluorescent dyes. (B), Graphical representation of (A). The cell number per high-power field (HPF, 6400)was the mean of the cell number obtained from 5 random HPF. Detached cells, which were stained with both annexin V and EtD-III positive, were alsoincluded in the cell number determination. Results are the mean 6 S.D. of 3 experiments.doi:10.1371/journal.pone.0010498.g003
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 7 May 2010 | Volume 5 | Issue 5 | e10498
decreases cell viability by blocking protein synthesis. Similar results
were obtained for Hela cells (data not shown).
Chick chorioallantoic membrane assayThe progression and metastasis of neoplastic disease requires
extensive vascularization of the tumor to sustain the nutritional and
oxygen demands of the proliferating cancer cells. This is generally
achieved through a process called angiogenesis [29]. The chick
chorioallantoic membrane assay (CAM) is a useful in vivo model
system to investigate the effects of toxins on angiogenesis. As shown in
Figure 6A, DT385 at a concentration of 40 nM (1.2 pmol in 30 mL)
caused complete inhibition of HEp3-induced (angiogenic) vascular
sprouting. This concentration is far below the IC50 of the tumor cells
and therefore a direct effect of DT385 on HEp3 viability is unlikely.
To investigate whether DT385 could reduce tumor growth, HEp3
tumors were grown on the CAM and the 6-day old tumors were then
treated with 2 mg of DT385 injected intravenously daily for three days.
Assuming that the total blood volume of a 15–17 day chick embryo is
approximately 2 mL [30], the initial systemic concentration of DT385
used in these experiments would be about 20 nM. Tumors were
removed and weighed 24 hours after the last injection. We observed
that DT385 decreased tumor mass by 41% (Figure 6B). We also
observed a significant loss of 66% in tumor volume after treatment
with DT385 (Figure 6C). These results indicate that DT385 can
inhibit tumor angiogenesis and subsequently reduce tumor growth.
Inhibition of tumor growth by DT385To study the effect of DT385 in another in vivo model, LLC were
implanted in mice and tumors were allowed to grow for a 19-day
period. DT385 was administered subcutaneously on days 5, 9, 12 and
15 at a dosage of 25 mg for first injection and 10 mg for the remaining
injections. We observed that tumor growth was significantly inhibited
in mice receiving DT385 by day 19 (23%) when compared to control
mice treated with an identical concentration of the recombinant
control protein, SUMO (Figure 7). We did not observe any toxicity to
the mice at these dosages. However, we observed an LD50 of 120 mg
of DT385 (n = 10, data not shown) suggesting that higher doses of
DT385 were lethal to mice.
Discussion
The mechanism by which DT enters cells has been well
established. The carboxyl-terminal receptor-binding domain (R
Figure 4. Internalization of DT385 by Cancer Cells. (A), Cancer cells were grown in an 8-well chamber slide. FITC-labeled DT385 (1 mM) wasadded to culture. The cells were observed and photographed under a Zeiss LSM 510 fluorescence confocal microscope (Carl Zeiss, Germany) after36 h. Top and right sides of images demonstrates the orthogonal view of the confocal dataset shown in the x and y-axis of images. Magnifications areindicated below images. (B), U87 cells were incubated with 2 mM DT385 alone or in combination with 10 mM NH4Cl for the indicated times. Themedia was then replaced and cells were cultured for a total time of 36 hours after which cell viability was accessed with the MTS assay. The PBSvehicle control was considered as 100% viable. Results are expressed as percentages of PBS-treated cells. Results shown are representative of 2independent experiments performed in triplicate.doi:10.1371/journal.pone.0010498.g004
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 8 May 2010 | Volume 5 | Issue 5 | e10498
domain-residues 391–535) interacts with the extracellular heparin-
binding epidermal growth factor-like growth factor precursor
protein and as a consequence the DT-receptor complex is
internalized [3,4]. Most cells possess this DT cell surface receptor
and therefore DT has the capability of killing a wide range of cell
types. However, it has been both assumed and also reported that
the removal of the R domain renders the resultant truncated
‘‘receptorless’’ DT incapable of binding to and being internalized
by cells. Previous studies had reported that the ‘‘receptorless’’
truncated DT was incapable of affecting cells [16]. Unexpectedly,
we observed that the ‘‘receptorless’’ truncated DT, DT385, is
capable of entering and killing a variety of cancer cells.
Interestingly, we observed that the efficacy of DT385 varied
among the different cell types. While some cancer cells such as U-
87 MG, U251, 293T, HEK293, Hela and Calu-3 cells had IC50
values less than 0.5 mM, other cancer cells Colo201, Colo205,
LNCap, PC-3, HT1080, and MDA-MB-231 cells has an
intermediate sensitivity to DT385 (IC50 between 0.5–1.5 mM).
Some cancer cell lines including MCF7, HCT116, BT-20, NB4,
HL-60, and HEp3 cells had IC50 values greater than 1.5 mM and
others were unaffected by DT385 at the concentrations examined.
Although speculative, it is possible that the diversity in efficacy to
DT385 demonstrated by cultured cells is due to either differential
rates of inactivation of DT385 by the cells or by differences in the
uptake of the toxin. We have shown that DT385 is internalized by
cells (Figure 4A), by an endocytotic mechanism (Figure 4B).
It was also interesting that primary cells and primary cell lines
were generally much more resistant to DT385 than cancer cells.
We initially considered the possibility that the efficacy of DT385
might be a function of the cell doubling time. However, this was
not true. For example, 293T, PC-3 and HCT116 are highly,
intermediately and weakly sensitive to DT385, respectively.
However, the growth rate of these cells is similar. The doubling
time is ,17.7 h for the HCT116 cells [31] ,18 for PC-3 cells
[32], and 22 h for 293T cells [33]. The glioma cell line U-87 MG
is very sensitive to DT385, whereas human endothelial cell line
Figure 5. Inhibition of protein synthesis by DT385. (A), U-87 MG cells were treated with 1 mM DT385 or control (either rP22, SUMO or PBS) for24, 36 h and 48 h, respectively, and then labeled with [35S-]methionine for 15 minutes. Cell lysates (10 mg) were separated by SDS-PAGE (12% gel)and gels were stained with Coomassie blue (left), dried on Whatman paper and visualized by radiography using a phosphorimager (right).Representative images for 36 h treatment are shown. (B), quantification of (A). Radioactivity of cell lysates was determined by liquid scintillationcounting. Data are expressed as percent of control. The average of CPM from irrelevant protein, rP22 or recombinant SUMO or PBS treatment wasconsidered as the control CPM value. Results are the mean 6 S.D. (n = 6, 2 independent experiments). Cell viability was also measured as described inthe legend to Figure 1. Results are the mean 6 S.D. of three independent experiments performed in triplicate.doi:10.1371/journal.pone.0010498.g005
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 9 May 2010 | Volume 5 | Issue 5 | e10498
HUVEC is not affected by DT385 up to 2.5 mM. These two cell
lines have the same doubling time of ,27 h [34,35]. We measured
the growth rate of 9 cell lines, but found no correlation between
the cell growth rate and the sensitivity to DT385 (Table 1). We
also observed that confluent cells were resistant to DT385.
Confluent cells are known to have greatly reduced rates of protein
synthesis [36]. Therefore, it is likely that the reduced rates of
protein synthesis observed with the confluent cells are responsible
for the ineffectiveness of DT385.
It is interesting to note that human endothelial cells are resistant
to DT-385 treatment in vitro but angiogenesis is inhibited in CAM
assays in vivo. The endothelial cells are not the only cells that are
involved in angiogenesis in the CAM system. The seemingly
conflicting data is possibly the result of an effect of DT-385 on
chicken inflammatory cells, as these are critical for a robust
angiogenic response. Alternatively, our observation that bovine
pulmonary artery endothelial cells are sensitive to DT385 suggests
that the species of the endothelial cell may be a factor in its
sensitivity to DT385.
It has been reported that one molecule of DT introduced into
the cytosol of a cell is sufficient to cause cell death [37]. Since one
DT molecule catalyses the ADP-ribosylation of 2000 EF-2
molecules/min in vitro [38], and since one cell contains about
106 EF-2 molecules [39], it can be roughly estimated that once a
single DT molecule has entered the cytoplasm it would require
about 30 h to inactivate all the EF-2 molecules in a cell [40].
Consistent with this calculation, we observed that the inhibition of
protein synthesis peaked 36 hours after exposure to DT385. We
also used FITC-labeled annexin V to detect apoptosis and
ethidium homodimer III (EtD-III) to examine the loss of cell
membrane integrity. We observed that maximal apoptosis and loss
of membrane integrity occurred after 72 h treatment (Figure S5).
The potential usefulness of DT385 as an antitumor agent was
evaluated in both the chick CAM assay (Figure 6) and the Lewis
lung carcinoma mouse model (Figure 7). Although, we observed a
significant reduction in tumor mass in both in vivo systems, the
reduction of the mouse tumors was not dramatic. However, the
mouse LLC cells used in these studies are among the least sensitive
Figure 6. DT385 inhibited angiogenesis and reduce tumor growth in a chick chorioallantoic membrane model. (A), HEp3 cells wereused to stimulate angiogenesis on the CAM. PBS control without HEp3 cells indicated the baseline level of angiogenesis. Angiogenesis, in thepresence of HEp3 cells and in the absence or presence of recombinant DT385 or recombinant control protein (SUMO) was analysed. Results are themean 6 S.E. of 80 data points from two replicate assays *p,0.05 (Student t test). (B), DT385 significantly decreased tumor growth. Hep3 tumorgrowth in the CAM system was assessed as described in Materials and Methods. Results are the mean 6 S.E. The mean values of tumors for DT385 orcontrol treatments were 13.8861.56 mg (mean 6 S.E.; n = 23) and 23.3364.3 mg, (mean 6 S.E.; n = 12) respectively, p = 0.012. (C), DT385 significantlydecreased tumor volume. Hep3 tumor volume in the CAM system was assessed as described in Materials and Methods. Results are the mean 6 S.E.The mean values of tumor size for DT385 or control treatments were 8.86610962.42 mm3 (mean 6 S.E.; n = 18) and 2.97610960.49 mm3 (mean 6S.E.; n = 8) respectively, p = 0.02.doi:10.1371/journal.pone.0010498.g006
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 10 May 2010 | Volume 5 | Issue 5 | e10498
cells to DT385. Extensive experimentation involving testing the
modality and frequency of injection of DT385, as well as the
concentration will be required before the potential efficacy of
DT385 as an antitumor agent can be fully appreciated. It is also
possible that combining DT385 with other chemotherapeutic
drugs may increase the efficacy of the toxin. Considering the high
sensitivity of glioma cell lines to DT385, we cannot rule out the
possibility that certain forms of cancer may be very sensitive to
DT385. For example, glioblastoma is form of astrocytic brain
tumors that is refractory to chemotherapy in most cases. We have
observed that U-87 MG cells, which are used as an in vitro model of
human glioblastoma are among the most sensitive cells to DT385
(Table 1). It is therefore possible that DT385 alone or in
combination with other anti-tumor drugs might prove effective
against this form of cancer. Taken together, our data directly
demonstrate that DT385 is a cytotoxin capable of killing tumor
cells and arresting tumor growth. Although we observed that
DT385 killed cultured cancer cells due to its ability to inhibit
protein synthesis, we cannot rule out the possibility that the toxic
effect of DT385 on tumor growth in our zenograft (Figure 6) and
ectopic models (Figure 7) was not due to protein synthesis
inhibition.
Our report has potentially important implications as to the
specificity of the fusion proteins that are currently being tested as
anticancer agents. It is currently believed that fusion proteins such
as DT388IL-3 and DT389-IL-2 (denileukin diftitox-Ontak)
recombinant toxin that are used in the treatment of lymphoma
derive their specificity by virtue of targeting the toxin to IL
receptors. However, our data suggests that at sufficiently high
concentration of toxin, cells lacking IL receptors may be non-
specifically targeted by the DT domain of the fusion protein.
Although at the low concentration of these toxins that are
currently deployed in treatment schemes, it is unlikely that non-
specific cytotoxic effects will occur, we cannot rule out the
possibility that at higher concentrations and with repeated doses of
these toxins, non-specific cytotoxic effects involving other cells or
tissues might occur.
Considering the prevalence of childhood vaccination for DT,
the truncated native protein might be subject to that established
immune response. While the native protein may not be the ideal
mechanism for clinical implementation, truncated DT could be
the basis for a functional anti-tumor strategy. Further modification
of the protein will be evaluated with the objective in mind of
minimizing its antigenicity and avoiding the established immune
protection in immunized individuals. Furthermore, ongoing work
in drug delivery systems such as liposomes may offer an effective
mechanism to shield truncated DT from the immune system while
on route to the tumor tissue.
The current strategy of intratumor injections provide an
effective proof of principle. Currently, intratumoral injections
have been shown to be useful for a limited number of tumors such
as malignant glioma. Future clinical implementation will require
systemic delivery. Fortunately, recent advances in the field of
cancer therapeutics will likely make it possible to package and
deliver anti-cancer therapeutics systemically.
Supporting Information
Figure S1 Effect of DT385 on human endothelial cells. A), Time
response curves. Cells were treated with 2.4 mM of either
recombinant p22, DT385 or DT385-p22, for 7 days. Viable cells
were determined at indicated time points as described in the
legend to Figure 1. Media were replaced every 3 days. B), Dose
response curves. The concentration of DT385 was increased to
10 mM, and cell viability was measured 72 h later as described in
the legend to Figure 1. Data are expressed as percent control
response versus incubation time. Results are the mean 6 S.D. of 3
experiments performed in triplicate.
Found at: doi:10.1371/journal.pone.0010498.s001 (0.17 MB TIF)
Figure S2 DT385 did not kill confluent cells. Cells were grew to
confluent first, then incubated in fresh media and treated with
2 mM of DT385 for 3 days. Viable cell numbers after the three day
incubation were quantified by the CellTiter 96 AQueous
(Promega) assay. Cells without treatments were used as controls.
Alternatively, PBS or control protein treatments were used as
controls. Data are expressed as percent control response. Results
are the mean 6 S.D. of 3 experiments performed in duplicate
(n = 6).
Found at: doi:10.1371/journal.pone.0010498.s002 (0.44 MB TIF)
Figure S3 Two Applications of DT385 and DT-p22 Increased
the Proliferation Inhibition Effect on Tumor Cells, but not on
Human Endothelial Cell Lines. Cells were first treated with
DT385 for 48 h, washed with PBS and then were further treated
with a second application of DT385 for 48h (2 treatments) or
simply incubated with DT385 for 4 days (1 treatment). Viable cells
were then quantified by the CellTiter 96 AQueous (Promega)
assay. A), Hep3, B), MDA-MB-231 cells were administrated
DT385 at the indicated concentration. C), HUVEC or HdMEC
cells were administrated 2 applications of DT385 at a concentra-
tion of 2.5 mM. PBS was used as a control treatment. Data are
expressed as percent of control treatments. Results are the mean 6
S.D of 2 experiments performed in triplicate (n = 6).
Found at: doi:10.1371/journal.pone.0010498.s003 (0.20 MB
TIF)
Figure S4 Internalization of DT385 by Cancer Cells. Cancer
cells were grown in an 8-well chamber slide. FITC-labeled DT385
or BSA (DT385-FITC or BSA-FITC, 1 mM) was added to culture.
The cells were observed and photographed with a Zeiss Axioplan
Figure 7. DT385 inhibited tumor growth in mice. Tumors wereinduced by subcutaneous injection of LLC cells (106 cells). RecombinantSUMO (control) or recombinant DT385 were administered to mice,peritumorally at day 5 (25 mg), and at days 9, 12 and 15 (10 mg eachinjection). Tumor weights with the DT385 or control (SUMO) treatmentswere evaluated as described in the Materials and Methods. The meanvalues of tumors for DT385 or control treatments were 1.8760.18 g(mean 6 S.E.; n = 8) and 2.4460.18 g (mean 6 S.E.; n = 5), respectively,p = 0.0264).doi:10.1371/journal.pone.0010498.g007
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 11 May 2010 | Volume 5 | Issue 5 | e10498
II fluorescence microscope (Carl Zeiss, Germany) after 36 h. A),
bright-field images. B), fluorescence images.
Found at: doi:10.1371/journal.pone.0010498.s004 (3.81 MB
TIF)
Figure S5 Time Course Assays of Apoptosis Caused by DT385.
U-87 MG cells growing in an 8- well chamber slide were treated
with 1.2 mM of DT385 (A) or control protein (B), respectively.
Following treatments, cells were stained for apoptosis with FITC-
labeled annexin V. Cell membrane integrity was evaluated with
ethidium homodimer III (EtD-III) according to the manufacturer’s
instructions (Biotium, Inc). Staining images were photographed
under a Zeiss Axioplan II fluorescence microscope (Carl Zeiss,
Germany). Digital images were processed in Adobe Photoshop
(Adobe Inc.). Representative images (1006 magnifications) were
shown.
Found at: doi:10.1371/journal.pone.0010498.s005 (7.69 MB
TIF)
Acknowledgments
We are grateful to Dr. S. Ramakrishnan for the DT385 construct. Glioma
cell lines U-87 MG and U251 were kindly provide by Dr. V. Wee Yong
(University of Calgary, Canada). The human epidermoid carcinoma cell
line HEp3 was a generous gift from Dr. Andries Zijlstra (Vanderbilt
University, USA). This manuscript was generated using open-source
software programs OpenOffice for the generation of the manuscript and
Zotero for the generation of the reference citations.
Author Contributions
Conceived and designed the experiments: YZ JDL DMW. Performed the
experiments: YZ WS DP KDP. Analyzed the data: YZ JDL DMW.
Contributed reagents/materials/analysis tools: AZ. Wrote the paper: YZ
AZ DMW.
References
1. Pappenheimer AM (1993) The story of a toxic protein, 1888-1992. Protein Sci 2:
292–8.2. Collier RJ (1975) Diphtheria toxin: mode of action and structure. Bacteriol Rev
39: 54–85.3. Naglich JG, Metherall JE, Russell DW, Eidels L (1992) Expression cloning of a
diphtheria toxin receptor: identity with a heparin-binding EGF-like growth
factor precursor. Cell 69: 1051–61.4. Iwamoto R, Higashiyama S, Mitamura T, Taniguchi N, Klagsbrun M, et al. (1994)
Heparin-binding EGF-like growth factor, which acts as the diphtheria toxin receptor,forms a complex with membrane protein DRAP27/CD9, which up-regulates
functional receptors and diphtheria toxin sensitivity. EMBO J 13: 2322–30.5. Madshus IH, Stenmark H (1992) Entry of ADP-ribosylating toxins into cells.
Curr Top Microbiol Immunol 175: 1–26.
6. FitzGerald D, Pastan I (1989) Targeted Toxin Therapy for the Treatment ofCancer. Curr Top Microbiol Immunol 81: 1455–1463.
7. Murphy JR, Bishai W, Borowski M, Miyanohara A, Boyd J, et al. (1986) Geneticconstruction, expression, and melanoma-selective cytotoxicity of a diphtheria
toxin-related alpha-melanocyte-stimulating hormone fusion protein. Proc Natl
Acad Sci U S A 83: 8258–62.8. Williams DP, Wen Z, Watson RS, Boyd J, Strom TB, et al. (1990) Cellular
processing of the interleukin-2 fusion toxin DAB486-IL-2 and efficient deliveryof diphtheria fragment A to the cytosol of target cells requires Arg194. J Biol
Chem 265: 20673–7.9. Chadwick DE, Williams DP, Niho Y, Murphy JR, Minden MD (1993)
Cytotoxicity of a recombinant diphtheria toxin-granulocyte colony-stimulating
factor fusion protein on human leukemic blast cells. Leuk Lymphoma 11: 249–62.10. Chan CH, Blazar BR, Greenfield L, Kreitman RJ, Vallera DA (1996) Reactivity
of murine cytokine fusion toxin, diphtheria toxin390-murine interleukin-3(DT390-mIL-3), with bone marrow progenitor cells. Blood 88: 1445–56.
11. Hotchkiss CE, Hall PD, Cline JM, Willingham MC, Kreitman RJ, et al. (1999)
Toxicology and pharmacokinetics of DTGM, a fusion toxin consisting of atruncated diphtheria toxin (DT388) linked to human granulocyte-macrophage
colony-stimulating factor, in cynomolgus monkeys. Toxicol Appl Pharmacol158: 152–60.
12. Hall PD, Willingham MC, Kreitman RJ, Frankel AE (1999) DT388-GM-CSF, anovel fusion toxin consisting of a truncated diphtheria toxin fused to human
granulocyte-macrophage colony-stimulating factor, prolongs host survival in a
SCID mouse model of acute myeloid leukemia. Leukemia 13: 629–33.13. Feuring-Buske M, Frankel A, Gerhard B, Hogge D (2000) Variable cytotoxicity of
diphtheria toxin 388-granulocyte-macrophage colony-stimulating factor fusionprotein for acute myelogenous leukemia stem cells. Exp Hematol 28: 1390–400.
14. Black JH, McCubrey JA, Willingham MC, Ramage J, Hogge DE, et al. (2003)
Diphtheria toxin-interleukin-3 fusion protein (DT(388)IL3) prolongs disease-freesurvival of leukemic immunocompromised mice. Leukemia 17: 155–9.
15. Cohen KA, Liu TF, Cline JM, Wagner JD, Hall PD, et al. (2004) Toxicologyand pharmacokinetics of DT388IL3, a fusion toxin consisting of a truncated
diphtheria toxin (DT388) linked to human interleukin 3 (IL3), in cynomolgusmonkeys. Leuk Lymphoma 45: 1647–56.
16. Ramakrishnan S, Olson T, Bautch V, Mohanraj D (1996) Vascular endothelial
growth factor-toxin conjugate specifically inhibits KDR/flk-1-positive endothelialcell proliferation in vitro and angiogenesis in vivo. Cancer Res 56: 1324–1330.
17. Wild R, Yokoyama Y, Dings R, Ramakrishnan S (2004) VEGF-DT385 toxinconjugate inhibits mammary adenocarcinoma development in a transgenic
mouse model of spontaneous tumorigenesis. Breast Cancer Res Treat 85:
161–171.18. Ramage JG, Vallera DA, Black JH, Aplan PD, Kees UR, et al. (2003) The
diphtheria toxin/urokinase fusion protein (DTAT) is selectively toxic to CD87expressing leukemic cells. Leuk Res 27: 79–84.
19. Urieto JO, Liu T, Black JH, Cohen KA, Hall PD, et al. (2004) Expression and
purification of the recombinant diphtheria fusion toxin DT388IL3 for phase Iclinical trials. Protein Expr Purif 33: 123–33.
20. Hogge DE, Yalcintepe L, Wong S, Gerhard B, Frankel AE (2006) Variantdiphtheria toxin-interleukin-3 fusion proteins with increased receptor affinity
have enhanced cytotoxicity against acute myeloid leukemia progenitors. Clin
Cancer Res 12: 1284–91.21. Kreitman RJ (2006) Immunotoxins for targeted cancer therapy. AAPS J 8:
E532–51.22. Turturro F (2007) Denileukin diftitox: a biotherapeutic paradigm shift in the
treatment of lymphoid-derived disorders. Expert Rev Anticancer Ther 7: 11–7.23. Duvic M, Talpur R (2008) Optimizing denileukin diftitox (Ontak) therapy.
Future Oncol 4: 457–69.
24. Lewis JD, Destito G, Zijlstra A, Gonzalez MJ, Quigley JP, et al. (2006) Viralnanoparticles as tools for intravital vascular imaging. Nat Med 12: 354–60.
25. Zijlstra A, Aimes RT, Zhu D, Regazzoni K, Kupriyanova T, et al. (2004)Collagenolysis-dependent angiogenesis mediated by matrix metalloproteinase-13
(collagenase-3). J Biol Chem 279: 27633–27645.
26. Zijlstra A, Lewis J, Degryse B, Stuhlmann H, Quigley JP (2008) The inhibition oftumor cell intravasation and subsequent metastasis via regulation of in vivo
tumor cell motility by the tetraspanin CD151. Cancer Cell 13: 221–234.27. Kwon M, Yoon C, Fitzpatrick S, Kassam G, Graham K, et al. (2001) p22 is a novel
plasminogen fragment with antiangiogenic activity. Biochemistry 40: 13246–13253.28. Honjo T, Nishizuka Y, Kato I, Hayaishi O (1971) Adenosine diphosphate
ribosylation of aminoacyl transferase II and inhibition of protein synthesis by
diphtheria toxin. J Biol Chem 246: 4251–60.29. Folkman J (1971) Tumor angiogenesis: therapeutic implications. N Engl J Med
285: 1182–1186.30. Kind C (1975) The development of the circulating blood volume of the chick
embryo. Anat Embryol 147: 127–132.
31. Li G, Nelsen C, Hendrickson EA (2002) Ku86 is essential in human somaticcells. Proc Natl Acad Sci USA 99: 832–837.
32. Gonzalez-Moreno O, Calvo A, Joshi BH, Abasolo I, Leland P, et al. (2005) Geneexpression profiling identifies IL-13 receptor alpha 2 chain as a therapeutic
target in prostate tumor cells overexpressing adrenomedullin. Int J Cancer 114:870–878.
33. Cherepanov P, Pluymers W, Claeys A, Proost P, De Clercq E, et al. (2000) High-
level expression of active HIV-1 integrase from a synthetic gene in human cells.FASEB J 14: 1389–1399.
34. Mercapide J, Lopez De Cicco R, Bassi DE, Castresana JS, Thomas G, et al.(2002) Inhibition of furin-mediated processing results in suppression of
astrocytoma cell growth and invasiveness. Clin Cancer Res 8: 1740–1746.
35. Bagley RG, Walter-Yohrling J, Cao X, Weber W, Simons B, et al. (2003)Endothelial precursor cells as a model of tumor endothelium: characterization
and comparison with mature endothelial cells. Cancer Res 63: 5866–5873.36. Benı́tez-King G, Cazares F, Meza I (1989) Synthesis and phosphorylation of
cytoskeletal proteins during the in vitro biogenesis of MDCK cell monolayers.J Cell Sci 93 (Pt1): 53–61.
37. Yamaizumi M, Mekada E, Uchida T, Okada Y (1978) One molecule of diphtheria
toxin fragment A introduced into a cell can kill the cell. Cell 15: 245–50.38. Moynihan MR, Pappenheimer AM (1981) Kinetics of adenosinediphosphor-
ibosylation of elongation factor 2 in cells exposed to diphtheria toxin. InfectImmun 32: 575–582.
39. Gill DM, Dinius LL (1973) The elongation factor 2 content of mammalian cells.
Assay method and relation to ribosome number. J Biol Chem 248: 654–658.40. Falnes PO, Olsnes S (1998) Modulation of the intracellular stability and toxicity
of diphtheria toxin through degradation by the N-end rule pathway. EMBO J17: 615–625.
DT385 Kills Cancer Cells
PLoS ONE | www.plosone.org 12 May 2010 | Volume 5 | Issue 5 | e10498